However I'd also point out that if that class, instead of being simply
'hidden', had been similar to Bootstrap's "sr-only", or even a more
fully spelled out "screen-reader-only", the later developer would have
been more likely to wonder "Hmm, maybe that's not simply hidden but
means something else, maybe I should try to look up or ask someone what
it means if I'm not sure"
Attempting self-documenting code always matters for successor
developers, not just in issues of accessibility. And labelling something
simply 'hidden' that is not fact always hidden is misleading your
I mean, in your example they left a comment with their thought process
-- the thing was labelled 'hidden' after all.
On 9/17/14 5:03 PM, Will Martin wrote:
> To digress a bit from LibGuides ...
> The biggest problem with accessibility is not technical: it's cultural.
> Producing HTML that meets basic accessibility tests is not all THAT
> difficult. The harder part is setting up a culture where everyone --
> everyone! -- who writes content for the web is trained on how to do it
> accessibly. A content editor who is clueless about accessibility can
> very easily screw up their pages without even knowing they're doing so.
> The same applies to developers. Once while reviewing a library site's
> code, I came across a chunk of HTML that looked like this (roughly):
> I don't know why this was here? It's invisible! Disabling.
> <a href="#top" class="hidden">Return to top</a>
> An earlier developer had put that in to assist screen reader users in
> getting back to the top of the page if they wanted. The "hidden" class
> was a correctly written class for hiding content while leaving it
> available for screen reader users. But the next person to fill that job
> wasn't trained on WHY and took it out again.
> If you really want to commit to accessibility, it needs to be a
> criterion in the job description for your developers, and there needs to
> be a training process in place for anyone who produces content for your
> site. Probably with refreshers at intervals.